Research on Forecast of Dynamic Measurement Error Based on GABP Neural Network

The BP Algorithm in training neural network has good local performance but it is easy to fall into local minimum, while the genetic algorithm has good global performance but it doesn't have the ability such as modifying in local space. Therefore the following combinatorial method is put forward. First of all, we train neural network based on genetic algorithm in global space. Then, we train the parameters of neural network using BP algorithm in local space. The model of training neural network using genetic algorithm is established, the mechanism of GABP neural network is introduced specially, and the genetic operators is designed. Finally the GABP neural network outperforms GA neural network in the aspect of dynamic error forecast is verified by computer emulation example, and related conclusion is given.